At a glance
| Supported data platforms | PostgreSQL (and AlloyDB, Aurora, Citus, Neon), Snowflake, Databricks SQL, BigQuery, Redshift, SQL Server, Fabric, Oracle, SQLite, Trino, Pandas, Spark, S3, Azure Blob, GCS |
| Anomaly detection | Not documented for GX Core |
| Freshness and volume monitoring | Through Expectations you define (no automatic monitoring) |
| Schema checks | Through column and schema Expectations you define |
| Validation rules | Yes, the core purpose: declarative Expectations |
| Lineage and root cause | Not documented |
| Alerting | Checkpoint Actions: Slack, Teams, email |
| Deployment | Self-run Python library (Linux or macOS); GX Cloud discontinued June 2026 |
| Licensing | Apache 2.0, free |
| Ownership | Open source project stewarded by Fivetran since May 2026 |
GX Core remains a solid free choice for Python teams that want expressive, code-defined data validation in their pipelines under an Apache 2.0 licence. Since GX Cloud is gone as a public product and stewardship passed to Fivetran in 2026, plan to run it yourself, and watch the project's release cadence and roadmap under its new steward.
Overview
GX Core is a Python library for data validation. You connect a Data Source (a SQL database, Spark or Pandas dataframe, or files in cloud storage), define Batches of data, and attach an Expectation Suite: a set of declarative assertions such as values not being null, falling in a range or matching a set. A Validation Definition pairs data with a suite, and a Checkpoint runs validations and triggers Actions such as Slack, Microsoft Teams or email notifications. Results can be rendered as Data Docs, human-readable HTML reports.
The project has been known as Great Expectations since its launch; the company rebranded the library as GX Core and its hosted service as GX Cloud. In May 2026 GX Cloud was acquired by FICO and withdrawn as a public product, and Fivetran became steward of GX Core, saying it would continue as an open source, community-driven project. The documentation lists officially supported sources and states that Airflow 2.9 or later is the only officially supported orchestrator, and that Windows is not currently supported.
Key features
- Declarative Expectations grouped in Expectation Suites, written in Python
- Validation Definitions and Checkpoints to run validations in pipelines
- Actions on results: Slack, Microsoft Teams and email notifications, plus custom Actions
- Data Docs: generated HTML documentation of Expectations and validation results
- Runs on Pandas, Spark and SQL databases including PostgreSQL, Snowflake, Databricks SQL, BigQuery, Redshift, SQL Server, Oracle and Trino
- Apache 2.0 licence with community-driven development
Integrations and ecosystem
Pros and cons
Pros
- Free under the Apache 2.0 licence
- Expressive, reusable validation rules in Python
- Covers dataframes and many SQL databases with one API
- Data Docs give readable validation reports for stakeholders
Cons
- No hosted option any more: GX Cloud was discontinued in June 2026
- No automatic anomaly detection or lineage; you write and maintain Expectations
- Windows is not officially supported, and Airflow is the only officially supported orchestrator
- Stewardship changed in 2026, so long-term roadmap depends on Fivetran and the community
Pricing
GX Core is free and open source under the Apache 2.0 licence. The hosted GX Cloud service was acquired by FICO and, according to the Great Expectations announcement of 6 May 2026, discontinued as a public product on 1 June 2026. The greatexpectations.io pricing and cloud pages still described GX Cloud Developer, Team and Enterprise options on 7 November 2026, which conflicts with that announcement, so do not treat GX Cloud as a purchasing option without confirmation.
Pricing checked on the vendor's official page on 7 November 2026. Prices change; confirm before buying.
Who it is best for
Python data engineers who want free, code-defined validation of data in pipelines across dataframes and SQL databases, and are comfortable running the tooling themselves.
Typical use cases
- Validate files or tables at each stage of an Airflow pipeline and stop the run on failure
- Check incoming partner data in Pandas or Spark before loading it into a warehouse
- Publish Data Docs so analysts can see which rules a dataset passed
- Assert value ranges and null rates on Snowflake or PostgreSQL tables after each load
Who should look elsewhere
- Teams that want automated monitoring without writing rules: see Monte Carlo or Metaplane.
- Teams that need a hosted service from the same project, since GX Cloud is no longer offered publicly: see Soda.
Verdict
GX Core is still one of the most capable open source data validation libraries, with a permissive licence and wide data source support. The 2026 changes (GX Cloud sold to FICO and discontinued, Fivetran as steward) make it a self-run library choice, so weigh the project's health under new stewardship alongside its features.
Frequently asked questions
Is Great Expectations still open source?
Yes. GX Core is licensed under Apache 2.0. Since May 2026 Fivetran is the steward of the project, which it says will continue as an open source, community-driven project.
What happened to GX Cloud?
Great Expectations announced on 6 May 2026 that FICO had acquired GX Cloud and that it would be discontinued as a public product on 1 June 2026. Some greatexpectations.io pages still describe GX Cloud plans, which conflicts with that announcement.
Is GX Core the same as Great Expectations?
Yes. GX Core is the current name of the open source Great Expectations library (installed as great-expectations from PyPI).
Which Python versions does GX Core support?
The documentation lists Python 3.10 to 3.14, with 3.15 and later not yet supported. Windows is not officially supported.
Sources
- An Update for the Great Expectations Community (6 May 2026)
- GX Core GitHub repository
- great-expectations on PyPI
- GX application integration support (documentation)
- GX Core overview (documentation)
- Great Expectations pricing page (still shows GX Cloud plans)
Checked November 2026.
Listed in: Data Quality & Observability.
How we research reviews and what star ratings mean: our editorial method.